Agent Skills
› OrchestratorInc/agent-orchestrator
› using-ao
using-ao
GitHub管理Orchestrator CLI,涵盖Worker、Session、Project、Cue及PR审查等全生命周期操作。支持会话控制、自动化调度、文件预览与报告持久化,是协调Agent工作流的核心工具。
Trigger Scenarios
需要启动或终止Agent Worker
管理Agent会话状态(如恢复、重命名)
执行PR合并或代码审查
配置项目或设置自动化任务
Install
npx skills add OrchestratorInc/agent-orchestrator --skill using-ao -g -y
SKILL.md
Frontmatter
{
"name": "using-ao",
"trigger": "Using the `ao` CLI, importing existing agent sessions, or managing Orchestrator sessions, projects, Cues, messages, and browser previews.",
"description": "Use the Orchestrator `ao` CLI to manage workers, sessions, projects, Cues, messages, browser previews, and daemon control. Also covers importing existing agent sessions."
}
Orchestrator CLI Catalog
ao is a thin CLI over the local Orchestrator daemon. Every command is ao <command> --help for the authoritative flag list.
| Command | What it does | When to use | Details |
|---|---|---|---|
spawn |
Spawn a project worker, orchestrator, or projectless standalone worker | Starting a new task or issue | commands/spawn.md |
session |
Manage agent sessions (list, kill, rename, restore, history, etc.) | Inspecting or controlling running/terminated sessions, or reading the full conversation after taking over from another agent | commands/session.md |
account |
List Claude Code and Codex subscription accounts, set the default, show or move a session's account | Starting work on a specific account, or moving a session off an account that hit its limit | commands/account.md |
agent |
Inspect installed harnesses and launch readiness | Choosing or diagnosing an agent harness | ao agent ls --refresh |
pr |
Merge a PR or resolve its review threads | Completing a reviewed change | ao pr merge / ao pr resolve-comments |
project |
Register, inspect, configure, or remove projects | Setting up or managing repos Orchestrator knows about | commands/project.md |
automation |
Manage durable recurring session automations | Creating, editing, disabling, deleting, and inspecting scheduled runs | commands/automation.md |
cue |
Create or list reusable project Cues | Saving a repetitive command or agent task at the user's request | commands/cue.md |
orchestrator |
List orchestrator sessions | Viewing which sessions are orchestrators | commands/orchestrator.md |
review |
Start, list, cancel, or submit Orchestrator's native reviewer for a worker's PR (default: this session) | Reviewing implementation PRs owned by workers. For someone else's PR, use a worker that claims it with automatic fixing off | commands/review.md |
send |
Send a message to a running agent session | Messaging a live agent; use --steer to correct it mid-turn |
commands/send.md |
report |
Persist a meaningful worker report | Checkpoints, blockers, decisions, outputs, and completion | commands/report.md |
artifact |
Register files in the artifact directory as deliverables | Making a finished file appear as one of the session's artifacts | commands/artifact.md |
preview |
Start a session-owned app or open an exact URL/file | Running and showing the worker's relevant app, Markdown, HTML, PDF, or image | commands/preview.md |
pane |
Split this session's tabs in the desktop window and run commands in its terminal tabs (Developer Mode preview) | Putting a terminal, file or reviewer beside the chat, or running a command in a terminal the user can see | commands/pane.md |
render |
Show a self-contained HTML page inline in a chat thread | Answering with a chart, table, diagram, or mockup | commands/render.md |
browser |
Inspect and control the session's shared live browser | Verifying a web app through snapshots, interactions, waits, screenshots, console, and errors | commands/browser.md |
browser (remote host) |
Drive the host's shared remote Chromium, used when AO_AGENT_BROWSER_URL is set |
Same browser verification on a self-hosted remote host; one browser is shared across sessions | commands/browser-remote.md |
remote-host |
Link this machine to the user's Orchestrator account, or turn its remote access on or off | Setting up a VM or other headless machine so the user's desktops and phones can use it | commands/remote-host.md |
start |
Fetch (if needed) and open the Orchestrator desktop app | Launching the app | commands/start.md |
stop |
Stop the Orchestrator daemon | Shutting down Orchestrator | commands/stop.md |
status |
Show daemon status | Verifying the daemon is up and healthy | commands/status.md |
triage |
Collect diagnostics and draft a problem report, even offline | Reporting a bug | commands/triage.md |
doctor |
Run local health checks | Diagnosing Orchestrator setup problems | commands/doctor.md |
import |
Import projects from a legacy Orchestrator install | Migrating from the old flat-file store | commands/import.md |
version |
Print version information | Checking installed version | - |
completion |
Generate shell completion scripts | Setting up tab completion | - |
Conventions
- For importing existing agent conversations from outside Orchestrator, read commands/import.md. This includes Claude Code and Codex terminal sessions.
- Most read commands accept
--jsonfor machine-readable output. -p / --projectscopes session subcommand lookups to one project.- Session and project ids are shown by
ao session lsandao project ls. - To refer to a session in Orchestrator Chat or the Orchestrator terminal, use the canonical in-app
link
ao://sessions/{project-id}/{session-id}. Read commands/session.md for identity, encoding, and safety rules before constructing one. - When the user asks to save a repeatable task as a Cue, use
ao cue list --jsonto check existing definitions, thenao cue create. An agent Cue stores a reusable instruction; a command Cue stores an exact shell command. This agent-facing command can create and read Cues, but cannot edit or delete them. Direct the user to project settings for changes to existing Cues. - When the user asks to bring a repository's Conductor settings (scripts, run actions, environment variables, prompt) into Orchestrator, read commands/conductor.md first.
--agentis an alias for--harnessonao spawn.- Every command accepts
-h / --helpfor the full flag list. - For frontend launch or preview selection, read
commands/preview.md before acting. Its static-file,
project-runtime, and artifact rules are load-bearing: artifacts render in the
Artifacts tab after
ao artifact add, never throughao previewunless the user explicitly asks to see one in the Browser panel. - For page inspection, interaction, or request diagnosis, read commands/browser.md. It defines shared-tab behavior and the opt-in network policy.
Use references.md only when a natural-language request does not map clearly to a command above.
Version History
- aa6e65d Current 2026-11-16 19:56


